Dear Acee, Chris and LSR WG,

In the Post Multi-Chassis era, it is more and more common to consider CLOS 
design for core router (aka De-segregated router). Unlike data center, it 
requires to run LSR in the “fabric” and dramatically increase the numbers of 
LSR speakers in the core network. So, the “Fabric” abstraction described in 
these drafts are important for deployment. We support the WG to work on them.

Best regards,
Yiu


On January 27, 2020 at 1:27:13 PM, Acee Lindem (acee) 
([email protected]<mailto:[email protected]>)

wrote:


Speaking as WG Co-chair:



At IETF 107, we had a protracted discussion of several drafts having  goal of 
reducing the amount of link-state information that must be flooded into the 
level-2 area. We have two drafts that do this essentially via abstraction of 
the level-1 areas. These are:



https://www.ietf.org/id/draft-li-lsr-isis-area-proxy-01.txt

https://www.ietf.org/id/draft-chen-isis-ttz-07.txt



There are various reasons why these drafts can’t consolidated involving both 
IPR and government restrictions. Refer to 
https://datatracker.ietf.org/meeting/106/materials/minutes-106-lsr-00 for the 
complete discussion.



We have another draft that also reduces the amount of link-state information 
each IS-IS router must maintain but using IS-IS reflectors. This is slightly 
different but also avoids leaking all the level-1 area link-state to the 
level-2 area.



https://www.ietf.org/id/draft-przygienda-lsr-flood-reflection-01.txt



Given the amount of overlap and the conflicts amongst these drafts, the 
chairs/Ads are now asking whether there is a really a strong requirement to 
advance one or more of these documents. Especially given that we are already 
moving forward with both IS-IS/OSPF flooding reductions and the Hierarchal 
IS-IS work. Additionally,  we anticipate we’ll reach an impasse in 
consolidating these drafts. We’d really like to hear from the operators that 
would deploy these mechanisms.



Thanks,

Acee and Chris

_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr

Reply via email to